![]() |
MeLOn
|
Variables | |
string | problem_name = "peaks" |
LOAD DATA ############################ enter data set information. More... | |
string | filename_data = "./data/peaks.csv" |
int | input_dim = 2 |
int | output_dim = 1 |
bool | scaleInput = True |
bool | normalizeOutput = True |
data = np.loadtxt(open(filename_data, "rb"), delimiter=",") | |
X = data[:, :-output_dim] | |
y = data[:, input_dim:] | |
X_norm = utils.scale(X, scaleInput) | |
y_norm = utils.normalize(y, normalizeOutput) | |
x_train | |
x_val | |
y_train | |
y_val | |
test_size | |
n_train = x_train.shape[0] | |
string | output_folder = "./data/Output/" |
SET PARAMETERS ############################ output filename. More... | |
string | filename_out = output_folder + problem_name |
list | network_layout = [10, 10] |
string | activation_function = 'relu' |
string | activation_function_out = 'linear' |
float | learning_rate = 0.001 |
kernel_regularizer = tf.keras.regularizers.l2(l=0.0001) | |
string | kernel_initializer = 'he_normal' |
string | optimizer = 'adam' |
int | epochs = 100 |
int | batch_size = 128 |
int | random_state = 1 |
model = tf.keras.Sequential() | |
BUILD MODEL ############################. More... | |
loss | |
metrics | |
training_time = time.time() | |
TRAINING ############################. More... | |
history | |
y_pred = model.predict(X_norm) | |
SAVE MODEL ############################. More... | |
string example_training_of_ANN.activation_function = 'relu' |
string example_training_of_ANN.activation_function_out = 'linear' |
int example_training_of_ANN.batch_size = 128 |
example_training_of_ANN.data = np.loadtxt(open(filename_data, "rb"), delimiter=",") |
int example_training_of_ANN.epochs = 100 |
string example_training_of_ANN.filename_data = "./data/peaks.csv" |
string example_training_of_ANN.filename_out = output_folder + problem_name |
example_training_of_ANN.history |
int example_training_of_ANN.input_dim = 2 |
string example_training_of_ANN.kernel_initializer = 'he_normal' |
example_training_of_ANN.kernel_regularizer = tf.keras.regularizers.l2(l=0.0001) |
float example_training_of_ANN.learning_rate = 0.001 |
example_training_of_ANN.loss |
example_training_of_ANN.metrics |
example_training_of_ANN.model = tf.keras.Sequential() |
BUILD MODEL ############################.
example_training_of_ANN.n_train = x_train.shape[0] |
list example_training_of_ANN.network_layout = [10, 10] |
bool example_training_of_ANN.normalizeOutput = True |
example_training_of_ANN.optimizer = 'adam' |
int example_training_of_ANN.output_dim = 1 |
string example_training_of_ANN.output_folder = "./data/Output/" |
SET PARAMETERS ############################ output filename.
string example_training_of_ANN.problem_name = "peaks" |
LOAD DATA ############################ enter data set information.
int example_training_of_ANN.random_state = 1 |
bool example_training_of_ANN.scaleInput = True |
example_training_of_ANN.test_size |
example_training_of_ANN.training_time = time.time() |
TRAINING ############################.
example_training_of_ANN.X = data[:, :-output_dim] |
example_training_of_ANN.X_norm = utils.scale(X, scaleInput) |
example_training_of_ANN.x_train |
example_training_of_ANN.x_val |
example_training_of_ANN.y = data[:, input_dim:] |
example_training_of_ANN.y_norm = utils.normalize(y, normalizeOutput) |
example_training_of_ANN.y_pred = model.predict(X_norm) |
SAVE MODEL ############################.
example_training_of_ANN.y_train |
example_training_of_ANN.y_val |